Prenatal cocaine exposure (PCE) can have broad effects on neural development and cognition, with language skills routinely being affected. Extant findings have revealed multiple deficits/delays in language function associated with PCE including: 1) delayed age for language acquisition 2) poorer receptive language 3) poorer expressive language (fewer words spoken) and 4) abnormal response to auditory stimuli. The specific pattern of findings observed indicates language processing deficits at multiple levels from low-level perceptual deficits (e.g., speech processing) to more metacognitive deficits (e.g., semantics, syntax, comprehension). What is unknown is whether these deficits stem from the same or different underlying neurobiological anomalies. Moreover, because extant studies have looked at infants and young children, the extent to which these profiles of language dysfunction change as system demands change, e.g., from phonological development (early childhood) to the understanding of complex semantic and syntactic relationships (late childhood and adolescence), is unknown. It is likely, given the teratogenic effect of PCE, that multiple systems are affected, but also that the observed impairments may differ as a function of changes in system demands associated with development and/or academic pressures. We hypothesize that the perceptual deficits observed are a result of anomalous development of auditory perceptual systems in the temporal lobe and that the higher-level language deficits observed are due to a more diffuse effect on prefrontal lobe function. Additionally, we predict a change in the relative weighting of the affected systems associated with development, which will result in an increase in metacognitve impairment for older children and adolescents. The proposed research will partner with an ongoing longitudinal study of PCE to examine the underlying neurobiology of language processing deficits in a well-characterized behavioral sample of PCE adolescents. Specifically, the goals of the proposed research are 1) to use event related potentials (ERP) to assess the neurocognitive language profiles of PCE adolescents and of control children who were not exposed to cocaine (NCE); 2) to examine the relationship between earlier behavioral performance (measured by standardized tests of language and cognitive skills) and these neurocognitive language profiles; and c) to link these brain-behavior patterns to polymorphisms in candidate genes that have been associated with perceptual or metacognitive skills (COMT and BDNF), in order to establish an endophenotype associated with poor language performance in PCE children. The research outlined in this proposal will have a direct impact our understanding of the relationship between prenatal cocaine exposure and the underlying neurobiology associated with impairments in critical language skills that are important for academic and social success.